Overlay Process Application to Non-Standard Picture Segments

Solution Overview

Problem

Existing video coding systems, such as AVC, HEVC, and VVC, lack the ability to apply overlay processes like film grain or denoising independently of slice and subpicture partitions, limiting flexibility in noise handling for non-standard areas within a picture, such as rectangular or non-rectangular regions, and areas with specific content or color values.

Innovation Solution

A method is introduced to decode overlay process parameters and picture partitioning parameters from the bitstream, allowing the overlay process to be applied to segment areas of the picture whose boundaries are not aligned with picture, subpicture, or slice boundaries, enabling flexible definition and application of overlay processes without relying on NAL units that describe picture partitioning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If overlay processes are applied per picture, per subpicture, or per slice using existing systems, then the overlay process can be applied to standard partition areas, but the flexibility to apply overlay processes to non-standard areas (rectangular areas not matching slice partitioning, non-rectangular regions, areas with specific content or color values) is limited

Engineering Contradiction:

Improveflexibility in applying overlay processes to different picture areasVSAvoidcomplexity of overlay process application system

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The picture is divided into multiple segment areas with different overlay process parameters. Each segment area can be independently defined with its own overlay characteristics, allowing flexible application to different regions without being constrained by traditional slice or subpicture boundaries. The segment areas are defined using syntax elements that specify their boundaries and associated overlay parameters.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different overlay process parameters are applied to different segment areas within the same picture. This allows each region to have locally optimized overlay characteristics tailored to its specific content requirements, such as applying stronger denoising to noisy regions while preserving details in important areas like faces or text.

Inventive Principle:

Principle #3Local quality

2Adaptability or versatility

If overlay processes are applied using existing slice and subpicture partitions, then the system maintains compatibility with standard video coding structures, but the ability to handle non-aligned boundaries and custom regions is restricted

Engineering Contradiction:

Improvecapability to handle non-standard picture areasVSAvoidprecision of overlay process application to specific areas

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The picture is divided into multiple segment areas with different overlay process parameters. Each segment area can be independently defined with its own overlay characteristics, allowing flexible application to different regions without being constrained by traditional slice or subpicture boundaries. The segment areas are defined using syntax elements that specify their boundaries and associated overlay parameters.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension of control by defining segment areas that are independent of the traditional slice and subpicture hierarchy. This allows overlay processes to be applied in a completely flexible manner, defining regions based on content requirements rather than coding structure constraints.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Ease of operation

If overlay processes are applied independently of slice and subpicture partitions, then flexibility is improved, but the complexity of bitstream processing and merging increases

Engineering Contradiction:

Improveease of applying overlay processes to arbitrary areasVSAvoidcomplexity of bitstream processing and merging

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The picture is divided into multiple segment areas with different overlay process parameters. Each segment area can be independently defined with its own overlay characteristics, allowing flexible application to different regions without being constrained by traditional slice or subpicture boundaries. The segment areas are defined using syntax elements that specify their boundaries and associated overlay parameters.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The segment area definition mechanism serves multiple functions: it defines regions for overlay process application, specifies boundaries independent of slice/subpicture structures, and provides a unified framework that works across different picture types and coding configurations. This multi-functional approach reduces the need for separate handling mechanisms.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20240292003A1Applying an overlay process to a picture

Publication Date: 2024.08.29 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

A method for applying an overlay process to a picture in a bitstream. In one embodiment the method includes decoding a first set of one or more overlay process parameters from syntax elements in the bitstream, the first set of one or more overlay process parameters specifying a first overlay process. The method also includes decoding a first set of one or more picture partitioning parameters from syntax elements in the bitstream, the first set of one or more picture partitioning parameters specifying a first segment area of the picture, wherein a boundary of the first segment area of the picture is not fully aligned with a boundary of the picture or a boundary of any subpictures of the picture or a boundary of any slices in the picture. The method further includes decoding the picture, wherein decoding the picture comprises applying the first overlay process on the first segment area of the picture using the first set of one or more overlay process parameters.